2020-07-30 –, Green Track
RANSAC.jl offers the Efficient RANSAC algorithm, a widely spread method to recognize simple primitives in point clouds of scanned objects. This talk will show that the package can not only be a basis for existing reconstruction processes, but a tool for further research as well.
We introduce the RANSAC.jl package that implements the Efficient RANSAC algorithm. It is a widely used tool in the field of digital shape reconstruction to recognize simple geometric primitives (plane, sphere, cylinder, torus, cone) in point clouds. This algorithm can not only be used alone, but also as part of complex reconstruction processes.
So far, mostly C++ implementations have been published, and to the best of my knowledge, this is the first one in Julia. The main goal of the implementation is establishing a flexible tool, while maintaining the same level of performance as existing solutions. This way, not only existing functionality is replicated, but also a new research tool is introduced.
Tamás Cserteg is a developer at SZTAKI EMI (Institute for Computer Science and Control, Research Laboratory on Engineering & Management Intelligence). He received his MSc in mechatronics engineering at BME (Hungary). During his studies, he was a demonstrator several times at different departments of the Faculty of Mechanical Engineering. He participated in numerous national competitions (organised for teams of engineering students), and he finished twice in the top five places. His main research interests are robotics - including human-robot collaboration - and digital shape reconstruction.